This plot shows Opportunities (running attempts + receiving attempts) plotted against fantasy points for the 2018 data. For this and other plots you can:
There is a clear linear relationship between fantasy points and opportunities (i.e. they are highly correlated). That means there is not a lot of new information to be gained by looking at opportunities in addition to fantasy points. This is most true for Tight Ends and Receivers, where the linear relationship is most clear.
However, it may be helpful to look at the outliers and determine if something will change that will give them more opportunities or increase their effectiveness this year. For example, Ezekiel Elliot had the most opportunities/game of any player by far in 2018, but he was only the sixth highest running back in terms of points/game. If his team has a better offensive line this year and he maintains the same number of opportunities, he could easily have the highest points/game this year. If you set the TDs/game slider to a lower maximum, you will see Elliot had by far the most pts/game of running backs that averaged less than .8 TDs/game in 2018.
This plot shows only the running backs. The line indicates the predicted pts/game based on opportunities. Players below the line underperformed (fewer points than they should have gotten considering how many opportunities they had) and players above the line overperformed. Since the top running backs are in a class of their own and do not fit the linear trend (much higher pts/game given their number of opportunities/game), I excluded them when fitting the trend line.
Similar to above plot, this one is only Tight Ends. The trend is much tighter but there are some outliers.
Similar to above plot, this one is only Receivers. The trend is much tighter but there are some outliers.
These tables show how the players’ actual points compare to the expected points based on opportunities. The “residual” is the difference between the predicted value (the value of the trend line) and the actual value. So players with extreme positive residuals are unusually far above the trend line, and players with extreme negative residuals are unusually far below the trend line.
I omitted the top running backs from this table, since they influence the trendline so much and make the model less accurate for average players.